This proposal aims to evaluate the role of small scale susceptibility variations as a mechanism of contrast in NMR imaging. Susceptibility effects are important modifiers of the NMR signal from heterogeneous samples. Examples include the air spaces in lung, the trabeculae of bones, the flow dependent effects of blood in capillaries, and the mode of action of one class of NMR contrast agents, so-called susceptibility agents, which includes superparamagnetic iron oxide particles and lanthanide chelates. We will study the physical factors that are important in susceptibility related contrast, and aim to understand the mechanism of action and design factors for contrast materials. We will quantify the geometrical factors that influence such effects on NMR images to better understand susceptibility related transverse relaxation in complex media. Novel methods for measuring such effects on the magnetic environment of tissue water, including the use of pulse gradient spin echo (PGSE) measurements of apparent diffusion and multiple quantum line shape studies, will be used to characterize and compare different media and the effects of different agents. A specific aim is to better understand the factors that affect the ability of intravascular and interstitial susceptibility agents to relax tissue volumes not directly accessed by the agents, and to evaluate the dependence of relaxation rate on concentration and distribution in different compartments. We aim to test the hypotheses that the relaxation rate per unit concentration of agent varies with tissue type and is modified by blood flow. The design factors and mechanisms of action of susceptibility agents will be evaluated by theoretical computer modelling of the effects of susceptibility variations on the apparent transverse relaxation time and diffusion of solvent molecules undergoing Brownian motion and coherent flow. The influence of particle size, motion, spatial arrangement, magnetic moment, field strength and pulse sequence parameters, on the observed relaxation effects in gradient and spin echo techniques, will be investigated both by simulation and experimentally. The effects of compartmentation will be explored by simulation and by studies using agents entrapped in liposomes and capillary phantoms. The relaxation rate changes observed in different media will be evaluated by T2 and T1 relaxometry at various field strengths. We will quantify the contributions from dipolar effects, from bulk susceptibility line broadening, and from diffusion of water among the field gradients set up inside the medium. Using PGSE and multiple quantum line shape measurements we will directly measure properties of the local fields experienced by water molecules. These parameters will be correlated with the agents composition and distribution in tissue. In rats, the contributions to signal reduction that arise from susceptibility differences between the intra-and extra- vascular spaces will be evaluated in different organs and compared to the short range dipolar effects of agents such as Gd-DTPA. In this way we anticipate being able to evaluate whether susceptibility agents will be suited as markers for blood flow or other purposes, and to highlight design features that are important.